Recent decades have witnessed an explosion of scientific research on emotion and decision making. This project aims to harness insights from this burgeoning body of research to develop a theoretically-grounded and empirically-tested foundation for improving the effectiveness of emotionally-evocative public service announcements (PSAs) that target smoking. PSAs constitute a cornerstone in federal- and state- funded cancer control efforts, but large gaps exist in understanding how and for whom emotionally-evocative PSAs are effective, revealing a critical need for research. Although anti-smoking PSAs can be effective at a population level, some evidence shows that they can also trigger unintended ?boomerang? effects. Moreover, results of previous studies on human emotions suggest that PSAs that induce sadness may ironically increase smokers? impatience for and valuation of cigarettes. Likewise, although many PSAs are designed to evoke feelings such as sadness, few, if any, systematic studies have taken a theory-guided and empirically rigorous approach to testing which specific emotions will trigger which behavioral effects. Research drawing on the widely-applied Appraisal Tendency Framework (Lerner & Keltner, 2000; 2001) provides a basis for such emotion specific effects. Applying this framework, this project will conduct three studies that compare the effects of sadness versus gratitude (and control emotions of similar valence to each) on immediate and long-term smoking behavior. We select sadness because our previous research reveals that sadness increases temporal discount rate (i.e., impatience), and contrast it with gratitude, which decreases temporal discount rate. We also select sadness because our formative research reveals that it is the most frequently evoked negative emotion in the Centers for Disease Control?s anti-tobacco public service campaign. In Study 1, we will assess the specific effects of sadness and gratitude on impatience for immediate smoking reward among a nationally representative web-based sample. To test for emotion specificity, we contrast sadness with disgust and gratitude with happiness. To understand for whom emotion induction yields greatest impact, we will conduct moderation analyses assessing social and demographic factors that predispose individuals to higher smoking risk. In Study 2a, we will assess the effect of sadness and gratitude on temporal discounting of immediate smoking reward, valuation of cigarettes, and smoking cognitions and behaviors. In Study 2b, we will assess the cumulative effect on cessation behavior of repeated exposure to emotion induction (x 4) over a 90-day period. The project will provide a theoretically-grounded framework for more precise prediction of the outcomes of exposure to PSAs. It will also provide a scientific base for future development of PSAs to reduce tobacco use and promote cancer prevention. Finally, because the project includes an interdisciplinary team representing public health, emotion science, decision science, economics, and statistics, the research integrates insights from multiple literatures.